DSANet: Dual-Branch Shape-Aware Network for Echocardiography Segmentation in Apical Views

Echocardiography is an essential examination for cardiac disease diagnosis, from which anatomical structures segmentation is the key to assessing various cardiac functions. However, the obscure boundaries and large shape deformations due to cardiac motion make it challenging to accurately identify the anatomical structures in echocardiography, especially for automatic segmentation. In this study, we propose a dual-branch shape-aware network (DSANet) to segment the left ventricle, left atrium, and myocardium from the echocardiography. Specifically, the elaborate dual-branch architecture integrating shape-aware modules boosts the corresponding feature representation and segmentation performance, which guides the model to explore shape priors and anatomical dependence using an anisotropic strip attention mechanism and cross-branch skip connections. Moreover, we develop a boundary-aware rectification module together with a boundary loss to regulate boundary consistency, adaptively rectifying the estimation errors nearby the ambiguous pixels. We evaluate our proposed method on the publicly available and in-house echocardiography dataset. Comparative experiments with other state-of-the-art methods demonstrate the superiority of DSANet, which suggests its potential in advancing echocardiography segmentation.

I. Introduction

Cardiovascular diseases (CVD) are the leading cause of health problems and death, affecting millions of patients worldwide [1]. Echocardiography (cardiovascular ultrasound) is one of the essential imaging tools in diagnosing heart diseases, such as heart failure, valvular disease, and other diseases, due to its low cost and good real-time performance [2]. The segmentation of the main sections of the heart, such as the left ventricle (LV), myocardium, and left atrium (LA), in echocardiographic images, plays a vital role in the calculations of clinical indices for an efficient diagnosis [3]. However, manual delineation in clinical practice is time-consuming and operator-subjective, adversely influencing the efficiency and accuracy of clinical parametric measurements [4]. Therefore, developing automatic and effective approaches is desirable for echocardiography segmentation, improving the work efficiency in clinical.